RBA rate cuts are fading to the horizon on inflation pressure, but the US Fed is expected to cut its policy rate twice more in this cycle, says Westpac's Luci Ellis

By Luci Ellis*

The US government shutdown has meant that some key data releases have not been published, particularly crucial labour market data. The data that are available from private sector sources have been mixed. With monetary policy still on the tight side and tariffs weighing on employment as well as lifting some prices, downside risks on the labour market have come to the fore.

Some observers are also pointing to possible downside risks for the US labour market from AI. The relative weakness in the graduate job market is seen in some quarters as evidence of the AI threat. In their current form, generative AI tools do best when replacing white-collar job tasks that are routine or easily standardised, such as building reports or compiling information. This kind of “donkey work” is typically given to the least experienced employees, ostensibly to help them learn. Weaker demand for labour for this reason is long-lasting, not ephemeral. Though in some sense structural, it would be another reason for the Fed to respond.

These downside risks and the ongoing uncertainty over government policy and functioning changes the calculus for US monetary policy. In our Market Outlook release for November, we added two further 25bp cuts to our forecast for Fed policy, to occur in the first half of the year. This is despite the ongoing inflation risks in the US, which we expect to be a binding constraint on further easing beyond the cumulative 50bp we have now pencilled in.

The outlook for RBA policy has shifted in the opposite direction. As we have detailed elsewhere, the higher September quarter inflation print has pushed out the feasible timing of future cash rate cuts. (And if inflation plays out according to the RBA’s forecasts rather than our own, then any cut in the cash rate will take even longer.)

Ordinarily, this shift in interest rate differentials should matter for exchange rates. The usual theory on exchange rate determination is that when interest rates rise in one country relative to another, assets denominated in the first country’s currency are more attractive to investors. Its exchange rate against the other country’s currency should therefore appreciate.

This is not what we have seen lately in AUD/USD. If anything, the AUD has moved sideways, against the USD and other major currencies like the euro since the middle of the year. There are a number of reasons for this.

The first consideration is that our forecasts are not always in line with market pricing, and it is market pricing for rates that feeds into the view for exchange rates. When we set out our view of Fed policy for 2025 at the beginning of the year, our view was far from consensus. Market pricing on the US rates outlook had been more dovish than our house view. Our early-2025 view has turned out to be closer to the actual outcome for the year than what was priced in at the time. However, this means that the additional downside in the rates outlook now emerging for 2026 is not such a departure from what the market had been pricing, just a lot later than it had been pricing.

The second consideration is that it is not just policy rates that matter for the views of participants in currency markets. While some models of AUD fundamentals have used policy rate differentials, it is more common to focus on longer maturities. For example, the RBA’s workhorse MARTIN model uses a two-year differential. There, the story is a bit different. At the 2-year and 3-year maturity, Australian government bond yields legged down in the wake of ‘Liberation Day’ in April and have been drifting up steadily since. US yields at that maturity fell less at the time, because the US was the source of the policy and tariff-related inflation that had changed the policy outlook there. They have since been drifting down, narrowing the spread to Australian yields; at this maturity, Australian yields are now higher than their US equivalents. Again, this would suggest appreciation pressure on the AUD relative to the USD, but not on the timing implied by shifting views on the monetary policy outlook.

More broadly, other asset returns and expectations also matter to asset allocation decisions across economies, especially equities, which then drives flow demand for particular currencies. These are currently working in the opposite direction to the pressures implied by interest differentials. Because the current AI boom is so focused on a few US-domiciled firms, shifting investor views of the sector and these firms find their expression in global flows into US equity markets and related infrastructure assets and thus the USD. The ebb and flow of investor concerns about US trade policy also manifests in relative asset demand and so exchange rates.

There are parallels here with the late 1990s and early 2000s ‘dot com’ boom. That boom, too, was highly concentrated in a tech sector predominantly located in the US or at least listed on the US stock market. Australia was seen as an ‘old economy’ and did not receive the same interest from investors as the US did. Accordingly, the AUD was noticeably undervalued then, at the same time the USD was almost as overvalued as it was at the beginning of this year.

At the beginning of this year, the USD hit peak overvaluation at the same time as ‘US exceptionalism’ narratives reached their zenith. These both unwound in the first half of the year, as we have previously detailed. More recently, as trade tensions have eased and deals have been done, some of the downside risk to US prospects have faded and been priced out of the USD exchange rate, which has therefore shown some resilience in recent weeks.

The USD is still overvalued on standard metrics, though. There are a range of ways to assess this, but a standard one is just the ‘real effective exchange rate’ – trade-weighted indices adjusted for relative inflation rates. These are available from a range of sources including the IMF and most central banks, including the RBA for the AUD.

Using the Fed’s “Broad Index”, the USD is about 15% above its longer-run average, compared with more than 20% earlier in the year. History would suggest that this overvaluation will correct itself over the course of several years. Investors want to be less long the USD in the wake of policy developments in the US. But they are also mindful that an overvalued exchange rate does tend to mean-revert over time, and therefore they want to avoid future capital losses relative to holding assets denominated in other currencies.

All this suggests that AUD should still appreciate against USD over the next few years. This is a USD story not a AUD story, though. And because so much hangs on sentiment, it could happen in fits and starts.

Luci Ellis the the Westpac Group chief economist. This commentary is reposted from here.